Human activity-aware coverage path planning for robot-based mosquito control

Abstract

Automating mosquito control is a pivotal advancement in the pest control industry with the primary objective of mitigating the prevalence of vector-borne diseases. Recent progress in pest control robotics has enabled the automation of mosquito activity restrictions. However, existing robotic solutions have exhibited limitations in effectively addressing mosquito control while lacking a sensitive strategy for maximizing area coverage with crowded areas as a priority. In response to these challenges, this article proposes a novel human-first approach for complete coverage path planning (HFA-CCPP) that leverages the Glasius Bio-inspired Neural Network (GBNN) to cover areas that simulate and consider human activity patterns systematically. In this study, a mosquito-capturing robot, Dragonfly, is demonstrated with HFA-CCPP. This article provides an in-depth exploration of the technical intricacies of the proposed solution. The efficacy of the proposed technique is evaluated in terms of total area coverage and times taken to cover the high human activity region in simulation and real-world environments by comparing results with traditional GBNN. Across all scenarios, the proposed HFA-CCPP surpasses the traditional method by delivering efficient area coverage with minimal time for human-dense area coverage and efficiency in mosquito trapping. This finding stands as a newfound direction in automated mosquito control, holding great potential for curbing vector-borne diseases.